Overview
You will join the Digital Twins in Healthcare group to develop probabilistic digital twin models for predicting thrombosis risk in cardiovascular diseases. The role supports precision cardiovascular care by integrating multi-modal patient data with population priors and scalable twins. You will build and benchmark surrogate models with quantified uncertainty to generalise across anatomies and populations. This is a full-time, fixed-term role tied to a 3-year EPSRC project, offering opportunities for professional development within a collaborative, hospital-adjacent research environment. You’ll contribute to impactful cardiovascular modelling at the intersection of engineering, AI, and data.
Pay / Benefits
- Professional development leave (at least 10 days per year pro-rata)
- London weighting
- Fixed-term contract for 3 years
- Collaboration with a leading research university and NHS hospital environment
Responsibilities
- Develop a digital-twin framework to model blood coagulation mechanisms in atrial fibrillation and coronary heart disease
- Develop and benchmark probabilistic kernel-based and deep-learning surrogates to infer time-dependent thrombosis biomarkers with quantified uncertainty
- Integrate the digital twin with probabilistic Bayesian frameworks and derive population-level priors from representative cohorts and primary care data
Key requirements
- PhD in biomedical engineering, computer science, applied mathematics, computational science, physics or related discipline
- Strong scientific programming skills in Python and/or C++
- Expertise in at least two: biophysical computational modelling (ANSYS), Gaussian-process emulators, deep learning models, neural/physics-informed models
- Experience with medical imaging data processing (e.g. segmentation)
- Ability to conduct independent quantitative research in a multidisciplinary team
- Independent discipline of research
- Ability to work effectively in multidisciplinary teams
- Strong communication of complex methods
- Python and/or C++ programming
- Biophysical computational modelling (e.g. ANSYS)
- Gaussian-process emulators
…